Alternate machinist level??

[jpfabricator]

I went to Harbor Freight today, and while roaming the isles, I had an idea come to me. In the absence of a machinist level, I wonder if one of the digital angle gauges would sufice in leveling a lathe?? Some paralells would be used to bridge the gap on the ways and the gauge would be calibrated to the table, floor ect.. It might not make the tool level with the world, but it would be level in its self. Since the gauge reads to the hundreth decimal place would this be ok?? Please feel free to coment, I would love the feed back.

 

[Richard King]

Put it on the cross-slide and crank it back and forth and adjust both ends so they read the same. If your not going to use coolant then leveling it plumb with the earth is not needed. After you get it as good as you can with it then take some test cuts using the 2 collar test we have documented in the Enco lathe forum. Give it a try and lets see. How much is the gage at HF?

 

[twstoerzinger]

My understanding on why to level machines is to insure that there is not deflection or warp being induced into the machine frame by the leveling feet or other adjustable members. This is especially true with lathe beds where it is easy to induce some twist into the bed by uneven loading of the feet. This can result in turning a taper rather than a cylinder. The precision level simply provides an easy means of determining that the machine member is not twisted or warped.

I did a quick look through the HF website, but did not find a digital level rated to .01 degree. Anyways, lets say that the HF electronic level is repeatable at .01 deg. In other words, from dead level, it registers .01 degrees from either side in a repeatable manner. 0.01 degrees is 36 seconds of arc which is about 0.002 inches per foot.

Starrett and other manufacturers like to rate their levels in seconds of arc per division.
When the bubble moves one division (mark on the glass) the level has moved the rated seconds of arc.
The standard 12 inch "machinist level" is usually rated at 80 - 90 seconds of arc which is about 0.005 inch per foot of run.
These levels are usually used in setting up a project on the machine - not necessarily to level the machine itself.
The HF level could do better at 36 seconds resolution.

The "master precision level" is usually what is used to level and align the machine itself.
The Starret version of the master level is rated at 10 seconds of arc per division which is about 1/2 thousandth per foot.
The resolution on the HF level would be about 3 times more sloppy.

For what its worth, there is a Chinese version of the master level also rated at 1/2 thousandth per foot. You can get a 12" version for about $100 and an 8" version for about $80. This is about 1/10 the cost of the S brand. Grizzly sells these and a few of the other importers. I have used the Starrett version when I was working. I bought an 8" Chinese version for home use just to see if it was any good. I have been impressed. I tested the China version on a level surface, then put a 0.001 feeler gauge under one extreme end. I tied this many times on both ends and found the level to be very repeatable at about 1-1/2 grads. You have to let the level stabilize for a minute or so. The heat from your hand has to dissipate before the level stabilizes. If you put a sheet of paper under one end, the bubble runs off the scale.

Terry S.

 

[jpfabricator]

I picked up the gauge for $29.99. My thinking is I can use the angle gauge for so much more. I will try it as a level, and use the feeler gauge test to see how acurate it is. Thanks for the feed back, keep it coming.

 

[Richard King]

I am curious if it works. I have aligned hundreds of lathes using a precision level in my 45+ years career as a machine tool rebuilder and on small machines I will first ask the owner if he is going to use coolant? If he says no I put an equal amount of weight on all the leveling screws that are resting in a leveling steel pad with a counter faced hole in them so the leveling bolt doesn't walk.

Many time on the short bed lathes I don't use a level, but only use the 2 collar test that I have linked to my first post. I have a Starrett Precision and no one that I know (accept a scientist or college professor) thinks in arc seconds. The level says on the plastic cover .0005" per 12". The Starrett precision level is 15" long by the way, so it can be confusing. I tell my students to use .00065" per division when leveling when using it.. I also have a King-Way Alignment Detector gage and a King-Way level that is .0003" per 12". If you holds the bubble on the line and not one division the lathe is better then needed. The spec we use for a conventional machine is .0002" per foot. So if he has a 36" lathe bed he could be off .0006" and be within new machine spec's. I say to wait 10 seconds for the bubble to rest or stop moving, though.

If the machine is worn then it might be more. I will sometimes twist the bed if it is worn so the worn machine will cut true on the OD. Another thing to do when using a machinist or precision level is to turn it 180 deg's and see if the bubble repeats to the same place. Terry your right on about the hand heat transfer, you also need to worry about sunlight and heat and AC blowers hitting the bubble and machine.

There is no absolute here when we are working with worn machines and as I said before JP wants to give it a try I think it's a great experiment and am very curious what he finds out. JP please take some photo and show us your results. Thanks. Rich

 

[itsme_Bernie]

Hey Richard-

My machine is visibly warn at the headstock end. My fingernail can just get under the wear on the front pyramid n my Heavy Ten.

You would left the back end foot to twist the bed?

Bernie

 

[Richard King]

When the machine is really worn bad there you can only twist it so much. I would not recommend twisting a bed and using pull downs and the leveling screws as you could put a permanent twist in the bed.. But you can twist it and test the leveling pads so you keep weight on all of them. Yes you would left the back left and the front right. The following link shows how you can twist the bed on a new machine, but it is the same principal on a worn machine. But remember if you twist so the OD will cut true you will screw up the ID bore.

LatheLevelingand2collartest.pdf

- - - Updated - - -

When the machine is really worn bad there you can only twist it so much. I would not recommend twisting a bed and using pull downs and the leveling screws as you could put a permanent twist in the bed.. But you can twist it and test the leveling pads so you keep weight on all of them. Yes you would left the back left and the front right. The following link shows how you can twist the bed on a new machine, but it is the same principal on a worn machine. But remember if you twist so the OD will cut true you will screw up the ID bore.

LatheLevelingand2collartest.pdf

 

[jpfabricator]

Sorry its taken so long to get back to this. I did some test on the Millrites knee. I used a 6" parallel that would just span the gap between the dovetail ways. I zeroed the unit out on the table while it was stuck on the parallel. If nothing else I know my table is parrallel to the knee).

I then inserted a .0025 feeler gauge under the right side, the angle gauge read .1 degree.
twstowerzinger, you were right. Because there is 2 decimal places I assumed it would read as such, but it does not.

I tryed a .0015 feeler gauge first but it did not read anything.
I removed the feeler gauge and the angle gauge went back to zero.
When I inserted the .0025 gauge back under the parallel, and leaned my weight against the knee the angle gauge would read another .1 degree. Sorry no pictures of this, as I didnt have enough hands.
Im not claiming that this method will replace a true machinist level, Im just giving people who are just getting started a plausible alternative to an expensive level. As I stated before, I will use the parallels, and the angle gauge a lot more than a precision level on other projects.
Hope this helps spmeone, God bless.